Systems and methods for decoding block and concatenated codes are provided. These include advanced iterative decoding techniques based on belief propagation algorithms, with particular advantages when applied to codes having higher density parity check matrices such as iterative soft-input soft-output and list decoding of convolutional codes, Reed-Solomon codes and BCH codes. Improvements are also provided for performing channel state information estimation including the use of optimum filter lengths based on channel selectivity and adaptive decision-directed channel estimation. These improvements enhance the performance of various communication systems and consumer electronics. Particular improvements are also provided for decoding HD radio signals, satellite radio signals, digital audio broadcasting (DAB) signals, digital audio broadcasting plus (DAB+) signals, digital video broadcasting-handheld (DVB-H) signals, digital video broadcasting-terrestrial (DVB-T) signals, world space system signals, terrestrial-digital multimedia broadcasting (T-DMB) signals, and China mobile multimedia broadcasting (CMMB) signals. These and other improvements enhance the decoding of different digital signals.
Legal claims defining the scope of protection. Each claim is shown in both the original legal language and a plain English translation.
1. A method for iteratively decoding a digital signal, the signal being an Orthogonal Frequency-Division Multiplexed (OFDM) signal comprising OFDM reference subcarriers, OFDM data subcarriers, and OFDM symbol intervals, the method comprising: a. performing OFDM demodulation and subcarrier de-mapping of the digital signal into system control and data sequence symbols to obtain received distorted modulated symbols; b. performing initial channel state information estimation based on the received distorted modulated symbols, wherein said received distorted modulated symbols are carried by a plurality of OFDM data subcarriers and a plurality of OFDM reference subcarriers in at least one OFDM symbol interval; c. performing phase correction of the received distorted modulated symbols based on the initial channel state information estimation into coherent demodulated signals; d. performing symbol-to-bit demapping of the coherent demodulated signals by calculating log-likelihood ratios of the system control and data sequences symbols; e. producing, by a decoder, one or more logical channel signals by de-interleaving and demultiplexing the log-likelihood ratios of the system control and data sequences symbols; f. processing the one or more logical channel signals by producing, by the decoder of step e., soft estimates of convolutional code coded bits using soft-input soft-output decoding of one or more convolutional codes associated with the one or more logical channels; g. calculating improved channel state information by performing at least one additional iteration of channel state information estimation based on at least some of said soft estimates of convolutional code coded bits; and h. performing enhanced convolutional decoding of a set of data associated with the digital signal using said improved channel state information.
2. The method of claim 1 , wherein the digital signal is selected from the group consisting of a hybrid digital radio signal, satellite radio signal, digital audio broadcasting (DAB) signal, digital audio broadcasting plus (DAB+) signal, digital video broadcasting-handheld (DVB-H) signal, digital video broadcasting-terrestrial (DVB-T) signal, world space system, terrestrial-digital multimedia broadcasting (T-DMB) signal, and China mobile multimedia broadcasting (CMMB) signal.
This invention relates to digital signal processing, specifically methods for handling various types of digital broadcast signals. The problem addressed is the need to efficiently process and utilize different digital broadcast signals, which may vary in format and transmission method. The invention provides a method for selecting and processing a digital signal from a predefined group of broadcast signal types. The supported signals include hybrid digital radio signals, satellite radio signals, digital audio broadcasting (DAB) signals, DAB+ signals, digital video broadcasting-handheld (DVB-H) signals, digital video broadcasting-terrestrial (DVB-T) signals, world space system signals, terrestrial-digital multimedia broadcasting (T-DMB) signals, and China mobile multimedia broadcasting (CMMB) signals. The method ensures compatibility with these diverse signal formats, allowing for seamless reception and processing across different broadcast standards. This approach enhances flexibility in digital broadcasting systems, enabling devices to adapt to multiple signal types without requiring separate hardware or software configurations for each format. The invention is particularly useful in consumer electronics, automotive systems, and portable devices where support for multiple broadcast standards is essential.
3. The method of claim 1 , wherein the initial channel state information estimation comprises: a. identifying symbols of the received distorted modulated symbols that are known pilot symbols and unknown symbols belonging to the reference subcarriers, and unknown data symbols belonging to the data subcarriers; b. estimating the unknown symbols; c. selecting a smoothing filter for time and frequency domain channel estimation; d. selecting filter lengths of said smoothing filter based on estimated time selectivity and frequency selectivity of said channel response wherein said time selectivity and frequency selectivity are estimated by using the known pilot symbols and at least one of the estimated unknown symbols; e. applying said selected smoothing filter to said received distorted modulated symbols; f. estimating the channel response of said filtered distorted modulated symbols, the estimation based on said known pilot symbols, estimated unknown symbols and a first set of unknown data symbols, the first set of unknown data symbols being selected as a fraction of the unknown data symbols having more reliability than the remaining fraction of the unknown data symbols; and g. with respect to symbol positions corresponding to said remaining fraction of the unknown data symbols, estimating the channel response by interpolation based on the estimated values of channel response obtained from said known pilot symbols, estimated unknown symbols and said first set of unknown data symbols.
This invention relates to wireless communication systems, specifically improving channel state information (CSI) estimation in orthogonal frequency-division multiplexing (OFDM) systems. The problem addressed is the accurate estimation of channel response in the presence of distortion, which is critical for reliable data demodulation. The method enhances initial CSI estimation by leveraging both known pilot symbols and unknown symbols to refine channel tracking. The process begins by identifying known pilot symbols and unknown symbols in the received distorted modulated symbols, including those in reference and data subcarriers. Unknown symbols are then estimated. A smoothing filter is selected for both time and frequency domain channel estimation, with filter lengths adjusted based on the estimated time and frequency selectivity of the channel. These selectivity metrics are derived using the known pilot symbols and at least one of the estimated unknown symbols. The selected smoothing filter is applied to the distorted symbols. The channel response is then estimated using the filtered symbols, incorporating known pilot symbols, estimated unknown symbols, and a subset of the most reliable unknown data symbols. For the remaining less reliable data symbols, the channel response is interpolated based on the previously estimated values from the pilot symbols, estimated unknown symbols, and the reliable data subset. This approach improves channel estimation accuracy by dynamically adapting to channel conditions and selectively utilizing high-reliability data symbols.
4. The method of claim 3 , wherein estimating the unknown symbols comprises: a. soft diversity combining of the system control data sequence symbols that carry the same symbol values; b. differentially decoding said soft diversity combined symbols to obtain a corresponding soft decoded control data sequence of bits; and c. reconstructing control data sequence symbols from said decoded control data sequence bits.
This invention relates to wireless communication systems, specifically improving the reliability of control data transmission in environments with interference or fading. The method enhances the accuracy of estimating unknown symbols in a received control data sequence by leveraging soft diversity combining, differential decoding, and symbol reconstruction. The process begins by performing soft diversity combining on system control data sequence symbols that carry identical symbol values. This step aggregates multiple received versions of the same symbol to improve signal quality and mitigate errors caused by channel distortions. The combined symbols are then differentially decoded to produce a soft decoded control data sequence of bits. Differential decoding compares adjacent symbols to extract the transmitted information, which is more robust against phase and amplitude variations in the channel. Finally, the decoded control data sequence bits are used to reconstruct the original control data sequence symbols, ensuring accurate recovery of the transmitted control information. This approach is particularly useful in scenarios where control data must be reliably transmitted despite challenging channel conditions, such as in wireless networks with multipath fading or interference. By combining diversity techniques with differential decoding, the method improves the resilience of control data transmission, reducing errors and enhancing system performance.
5. The method of claim 3 , wherein estimating the unknown symbols comprises: a. soft diversity combining of the system control data sequence symbols that carry the same symbol values; b. differentially decoding said soft diversity combined symbols to obtain a corresponding soft decoded control data sequence of bits; c. decoding said soft decoded control data sequence bits to obtain improved decoded control data sequence bits using single parity check code bits wherein the soft decoded control data sequence bit comprises a parity bit protected field of two or more bits, and upon determining that the parity does not check, flipping a least reliable soft decoded control data sequence bit in said parity bit protected field; and d. reconstructing control data sequence symbols from said improved decoded control data sequence bits.
This invention relates to wireless communication systems, specifically improving the reliability of control data sequence decoding in the presence of interference or noise. The method addresses the challenge of accurately recovering control data symbols when transmitted signals are corrupted, which is critical for maintaining communication integrity in harsh environments. The process begins by performing soft diversity combining on system control data sequence symbols that carry identical values. This step enhances the signal quality by leveraging multiple received versions of the same symbol. The combined symbols are then differentially decoded to produce a soft decoded control data sequence of bits, which retains reliability information from the combining step. Next, the soft decoded bits are further processed using single parity check code bits. The decoded sequence includes a parity-protected field of two or more bits. If the parity check fails, the least reliable bit within this field is flipped to correct the error. This adaptive error correction improves the accuracy of the decoded control data sequence. Finally, the improved decoded bits are reconstructed into control data sequence symbols, resulting in a more reliable output. This method ensures robust control data recovery even under adverse conditions, enhancing system performance in wireless communication networks.
6. The method of claim 1 , wherein said at least one additional iteration of channel state information estimation comprises: estimating a channel response; selecting a smoothing filter for time and frequency domain channel estimation; and selecting filter lengths of said smoothing filter based on estimated time selectivity or frequency selectivity of said channel response.
This invention relates to wireless communication systems, specifically improving channel state information (CSI) estimation in environments with time-varying or frequency-selective channels. The problem addressed is the degradation of CSI accuracy due to channel variations, which impacts data transmission reliability and efficiency. The method involves performing multiple iterations of CSI estimation to refine accuracy. In each iteration, a channel response is estimated to assess the channel's behavior. Based on this response, a smoothing filter is selected for both time and frequency domain channel estimation. The filter lengths are dynamically adjusted according to the estimated time selectivity (how quickly the channel changes over time) or frequency selectivity (how the channel varies across frequencies). This adaptive filtering helps mitigate distortions caused by multipath fading or Doppler effects, improving signal quality. The technique ensures that the smoothing filter is optimized for the specific channel conditions, balancing between over-smoothing (which can obscure rapid changes) and under-smoothing (which may fail to reduce noise). By tailoring the filter parameters to the channel's characteristics, the method enhances the precision of CSI, leading to better adaptive modulation, beamforming, and resource allocation in wireless systems. This approach is particularly useful in high-mobility scenarios or environments with significant multipath interference.
7. The method of claim 6 , wherein at least one of the filter lengths selected during said additional iteration of channel state information estimation is shorter than at least one of the estimation filter lengths used in the initial channel state information estimation.
This invention relates to wireless communication systems, specifically improving channel state information (CSI) estimation accuracy in multi-antenna environments. The problem addressed is the computational complexity and accuracy trade-offs in CSI estimation, where longer filter lengths improve accuracy but increase processing overhead, while shorter filters reduce overhead but may degrade performance. The method involves an iterative CSI estimation process. Initially, CSI is estimated using a set of estimation filter lengths. If the initial estimation does not meet a predefined accuracy threshold, an additional iteration is performed. During this iteration, at least one of the filter lengths is reduced compared to the initial estimation. This adaptive approach balances accuracy and computational efficiency by dynamically adjusting filter lengths based on estimation quality. The shorter filters in subsequent iterations reduce processing overhead while still refining the CSI estimate. The technique is particularly useful in systems like massive MIMO or beamforming, where precise channel knowledge is critical but computational resources are limited. By selectively shortening filter lengths in later iterations, the method achieves a more efficient estimation process without sacrificing accuracy. The invention ensures reliable communication performance while optimizing resource usage.
8. The method of claim 1 , wherein said soft estimates of convolutional code coded bits are log-likelihood ratios of said convolutional code coded bits.
The invention relates to error correction in digital communication systems, specifically improving the decoding of convolutional codes. Convolutional codes are widely used in wireless and wired communications to detect and correct errors introduced during transmission. A key challenge in decoding these codes is accurately estimating the likelihood of transmitted bits, which is essential for reliable error correction. The invention addresses this challenge by refining the estimation process. Specifically, it involves generating soft estimates of convolutional code-coded bits, where these estimates are expressed as log-likelihood ratios (LLRs). LLRs provide a probabilistic measure of the likelihood that a transmitted bit is either 0 or 1, which is crucial for iterative decoding algorithms like the Viterbi or BCJR algorithms. By using LLRs, the decoding process becomes more robust, as it leverages probabilistic information rather than hard decisions, leading to improved error correction performance. The method integrates with existing convolutional decoding frameworks, enhancing their accuracy without requiring significant modifications to the underlying decoding logic. This approach is particularly beneficial in high-noise environments or applications where low error rates are critical, such as in 5G networks, satellite communications, or deep-space data transmission. The use of LLRs ensures that the decoding process is both efficient and reliable, making it suitable for a wide range of communication systems.
9. The method of claim 1 , wherein said soft-input soft-output decoding of said one or more convolutional codes is based on a Log-MAP algorithm.
A method for decoding convolutional codes in communication systems, particularly in scenarios requiring high reliability and efficiency, such as wireless or digital transmission. The method addresses the challenge of accurately decoding convolutional codes while minimizing computational complexity and error rates. It employs a soft-input soft-output (SISO) decoding technique, which processes both the received signal and its reliability information to improve decoding accuracy. The SISO decoder uses a Log-MAP (Maximum A Posteriori Probability) algorithm, which is a computationally efficient variant of the MAP algorithm. The Log-MAP algorithm avoids numerical instability issues by operating in the logarithmic domain, making it suitable for practical implementations. The method can be applied to multiple convolutional codes, enhancing performance in systems where multiple data streams or channels are decoded simultaneously. The use of Log-MAP ensures robust decoding with reduced complexity compared to traditional MAP decoding, making it ideal for real-time applications. The method is particularly useful in error-prone environments where reliable data recovery is critical.
10. The method of claim 1 , wherein said soft-input soft-output decoding of said one or more convolutional codes is based on a Max-Log-MAP algorithm.
This invention relates to error correction in digital communication systems, specifically improving the decoding of convolutional codes using soft-input soft-output (SISO) techniques. Convolutional codes are widely used in wireless and wired communications to detect and correct errors introduced during transmission. Traditional decoding methods, such as the Viterbi algorithm, provide hard-decision outputs, which may not fully leverage the reliability information (soft information) available in the received signal. The invention addresses this limitation by employing a Max-Log-MAP (Maximum Log-Maximum A Posteriori) algorithm for SISO decoding. The Max-Log-MAP algorithm is a computationally efficient approximation of the Log-MAP algorithm, which itself is an approximation of the optimal MAP algorithm. By using this approach, the invention enhances decoding accuracy while maintaining low computational complexity. The method involves processing the received signal to generate soft reliability metrics for each decoded bit, which are then used to improve subsequent decoding stages or iterative decoding processes. This technique is particularly useful in systems requiring high reliability, such as modern wireless standards like 5G, where iterative decoding and soft information exchange between components are common. The invention ensures robust error correction while optimizing computational efficiency, making it suitable for real-time applications.
11. The method of claim 1 , wherein said enhanced decoding of the set of data associated with the digital signal comprises at least one of list soft-input soft-output convolutional decoding and soft-input soft-output Cyclic Redundancy Check (CRC) decoding, and wherein the list soft-input soft-output convolutional decoding produces a sequence of soft convolutional code information bits and a list that includes at least some of most likely sequences of hard decisions of convolutional code information bits.
This invention relates to digital signal processing, specifically methods for enhancing the decoding of data associated with digital signals. The problem addressed is improving the accuracy and reliability of decoding processes in digital communication systems, where errors in transmitted data can occur due to noise or interference. The method involves enhanced decoding of a set of data associated with a digital signal. This enhanced decoding includes at least one of two techniques: list soft-input soft-output convolutional decoding and soft-input soft-output Cyclic Redundancy Check (CRC) decoding. The list soft-input soft-output convolutional decoding produces a sequence of soft convolutional code information bits, which are probabilistic representations of the decoded bits. Additionally, it generates a list of the most likely sequences of hard decisions of convolutional code information bits, where hard decisions are definitive binary values (0 or 1) derived from the soft bits. This list helps in selecting the most probable correct sequence, improving decoding accuracy. The soft-input soft-output CRC decoding further refines the decoded data by leveraging CRC checks to detect and correct errors, ensuring data integrity. The combination of these techniques enhances the overall robustness of the decoding process in digital communication systems.
12. The method of claim 11 , wherein said list soft-input soft-output convolutional decoding is selected from the group consisting of a list Log-MAP decoding, list Max-Log-MAP decoding, and list soft-output Viterbi decoding.
This invention relates to error correction in digital communication systems, specifically improving the performance of convolutional decoding algorithms. Convolutional codes are widely used to detect and correct errors in transmitted data, but traditional decoding methods like Viterbi or MAP decoding have limitations in terms of accuracy and computational efficiency. The invention addresses these issues by implementing a list-based soft-input soft-output (SISO) convolutional decoding approach. This method enhances decoding accuracy by considering multiple candidate paths during the decoding process, rather than relying on a single path as in conventional techniques. The invention specifies three particular decoding algorithms that can be used: list Log-MAP decoding, list Max-Log-MAP decoding, and list soft-output Viterbi decoding. Each of these algorithms operates by maintaining a list of the most likely sequences of decoded bits, allowing for more robust error correction. The list-based approach improves performance in noisy communication channels by reducing the likelihood of incorrect decisions. The invention is particularly useful in applications where high reliability is critical, such as wireless communications, satellite transmissions, and data storage systems. By selecting the appropriate list-based decoding method, the system can balance computational complexity with decoding accuracy, making it adaptable to different operational constraints.
13. The method of claim 1 , wherein enhanced decoding of the set of data associated with the digital signal comprises: a. list soft-input soft-output decoding of a convolutional code to produce a sequence of soft convolutional code information bits and a list that includes at least some of most likely sequences of hard decisions of information bits outputs of the convolutional code; b. upon determining that any of hard decision sequences of said soft information bits outputs of the convolutional code do not correspond to a codeword, soft-input soft-output decoding of an outer code using said soft information bits outputs of the convolutional code; and c. upon determining that a hard decision sequence of said soft information bits outputs of the convolutional code corresponds to a codeword of the outer code, using the hard decision sequence as a decoded signal.
This invention relates to digital signal processing, specifically methods for enhancing the decoding of data associated with digital signals. The problem addressed is improving the reliability and accuracy of decoding in communication systems where digital signals are encoded with convolutional and outer codes to protect against errors. The method involves a multi-stage decoding process. First, a convolutional code is decoded using a list soft-input soft-output (SISO) decoding technique, producing both soft convolutional code information bits and a list of the most likely hard decision sequences of information bits. If none of these hard decision sequences match a valid codeword, the method proceeds to decode an outer code using the soft information bits from the convolutional code. If a hard decision sequence from the convolutional code matches a valid codeword of the outer code, that sequence is used as the final decoded signal. This approach ensures robust error correction by leveraging both inner and outer code decoding stages, improving overall system performance in noisy or error-prone environments. The technique is particularly useful in applications requiring high reliability, such as wireless communications, data storage, and error-critical transmission systems.
14. The method of claim 1 , wherein enhanced decoding of the set of data associated with the digital signal comprises: a. one of a soft-input soft-output decoding of a convolutional code and a list soft-input soft-output decoding of a convolutional code; and b. soft-input soft-output decoding of an outer code represented by a parity check matrix.
This invention relates to digital signal processing, specifically methods for enhancing the decoding of data encoded with convolutional and outer codes. The problem addressed is improving the reliability and accuracy of decoding in communication systems where data is transmitted as digital signals, often subject to noise and interference. The method involves decoding a set of data associated with a digital signal using a combination of soft-input soft-output (SISO) decoding techniques. For convolutional codes, the method employs either a traditional SISO decoder or a list SISO decoder, which provides multiple candidate outputs to improve decoding accuracy. Additionally, the method includes SISO decoding of an outer code, which is represented by a parity check matrix. This outer code may be a low-density parity-check (LDPC) code or another type of code that benefits from parity-based decoding. By combining these decoding techniques, the method enhances the overall decoding performance, reducing errors and improving data recovery in noisy environments. The approach is particularly useful in applications such as wireless communications, data storage, and error correction systems where robust decoding is critical. The use of list decoding for convolutional codes and SISO decoding for outer codes allows for more flexible and efficient error correction, adapting to different signal conditions and code structures.
15. The method of claim 14 , wherein said soft-input soft-output decoding of said outer code operates on log-likelihood ratios of information bits outputs of the convolutional code using message passing and comprises: a. generating P>1 binary parity check matrices with at least L sparse columns, wherein L is the number of rows of the parity check matrix and up to L sparse columns contain only a single entry equal to 1 per column, wherein sparse columns of each of P parity check matrices correspond to different subsets of up to L bit log-likelihood ratios of L+R least reliable bit log-likelihood ratios, wherein R≧P is a configurable integer; b. iteratively decoding said log-likelihood ratios using said P parity check matrices with sparse columns to produce updated log-likelihood ratios and employing soft-input soft-output message passing decoding, wherein the number of iterations of said decoding is based on one of a predetermined number of iterations and a determination that at least one of said P matrices produces a valid codeword; and c. upon determining that decoding using said P parity check matrices produces no valid codeword, performing additional decoding based at least in part on algebraic decoding of the sequences of said updated log-likelihood ratios.
This invention relates to error correction in digital communication systems, specifically improving soft-input soft-output (SISO) decoding of outer codes in concatenated coding schemes. The problem addressed is enhancing decoding reliability and efficiency when processing log-likelihood ratios (LLRs) of information bits from an inner convolutional code. The method involves generating multiple (P>1) binary parity check matrices, each with at least L sparse columns, where L is the number of rows in the matrix. Each sparse column contains only a single entry equal to 1, and these sparse columns correspond to different subsets of up to L bit LLRs from the L+R least reliable LLRs, with R≥P being a configurable integer. The LLRs are then iteratively decoded using these P parity check matrices with sparse columns, employing SISO message passing decoding. The decoding process continues for a predetermined number of iterations or until at least one matrix produces a valid codeword. If no valid codeword is found, additional decoding is performed, incorporating algebraic decoding of the sequences of updated LLRs to further improve reliability. This approach leverages multiple parity check matrices and adaptive decoding strategies to enhance error correction performance in concatenated coding systems.
16. The method of claim 15 including storing updated P sequences of bit log-likelihood ratios obtained after IT≧1 iterations of said soft-input soft-output message passing decoding using P parity check matrices and on the condition that no valid codeword is produced using said additional decoding, further comprising the following steps: a. identifying Q≦L disagreement positions on hard decisions of said P sequences of bit log-likelihood ratios; b. generating P new parity check matrices such that the sparse columns include L columns corresponding to said Q≦L disagreement positions and, on the condition that Q<L, up to L-Q sparse columns corresponding to different combinations of L-Q of L-Q+R least reliable said updated P sequences of bit log-likelihood ratios, wherein R≧P is the configurable integer; c. decoding said updated log-likelihood ratios using said P new parity check matrices to produce further updated log-likelihood ratios, employing soft-input soft-output message passing decoding until a desired number of iterations is reached or until decoding using at least one of said P new parity check matrices produces a valid codeword; and d. on the condition that said decoding using said P parity check matrices produces no valid codeword, performing additional decoding that is based at least in part on algebraic decoding of the sequences of said further updated log-likelihood ratios.
This invention relates to error correction in digital communication systems, specifically improving the reliability of low-density parity-check (LDPC) decoding when initial decoding attempts fail. The problem addressed is the inefficiency of conventional LDPC decoding methods when they fail to converge to a valid codeword, leading to wasted computational resources and potential data loss. The method involves iterative soft-input soft-output message passing decoding using multiple parity check matrices. If no valid codeword is produced after a threshold number of iterations, the method proceeds with additional decoding steps. First, it identifies up to L disagreement positions in the hard decisions of the updated bit log-likelihood ratios (LLRs) from the previous decoding steps. Next, it generates new parity check matrices that include columns corresponding to these disagreement positions and, if fewer than L disagreements are found, additional columns based on the least reliable LLRs from the updated sequences. The method then performs further decoding using these new matrices until either a valid codeword is found or a maximum iteration count is reached. If decoding still fails, an algebraic decoding step is applied to the further updated LLRs to improve the likelihood of successful recovery. This approach enhances decoding reliability by dynamically adapting the parity check matrices based on disagreement positions and reliability metrics, reducing the probability of decoding failure.
17. The method of claim 15 , wherein message passing is performed using best graph belief propagation.
A system and method for optimizing message passing in a network of interconnected nodes, such as in distributed computing or sensor networks, addresses the challenge of efficiently exchanging information while minimizing computational overhead and latency. The system involves a network of nodes where each node communicates with neighboring nodes to propagate data or beliefs across the network. The method includes determining a message passing scheme that optimizes the flow of information, ensuring that messages are transmitted in a manner that reduces redundancy and improves convergence to a stable state. In one implementation, the message passing is performed using best graph belief propagation, a technique that leverages probabilistic graphical models to iteratively update beliefs at each node based on incoming messages from neighboring nodes. This approach enhances the accuracy and efficiency of information propagation by prioritizing the most relevant or reliable messages, thereby improving the overall performance of the network. The system may also include mechanisms for dynamically adjusting the message passing parameters based on network conditions, such as node failures or changes in connectivity, to maintain robustness and reliability. The method is particularly useful in applications requiring real-time data processing, such as distributed sensor networks, decentralized optimization, and machine learning systems.
18. The method of claim 14 , wherein said soft-input soft-output decoding of said outer code comprises simple greedy scheduling of check equation updates, for at least M equations, where 1<M<L, during an iteration in decoding of the outer code represented by a parity check matrix with at least L parity check rows, the method further comprising: a. for the set of M′<M of non-updated check nodes out of M check nodes of the parity check matrix, calculating values Val i =Min 1 +Min 2 , i=1, . . . , L, where 1<L<M′ and where Min i and Min 2 are the two smallest values in the set of absolute values of variable-to-check messages {|Mvc(i,:)|} where index i corresponds to the set of non-updated check nodes; b. sorting the calculated set {Val i } in decreasing order as an ordering vector I={I i , I 2 , . . . , I L }, such that I 1 is the index of a check node with largest value Val, I 2 is the index of a check node with the next largest value Val and I L is the index of a check node with the smallest value Val; c. updating L check node equations according to the ordering vector, I={I 1 , I 2 , . . . , I L } calculated in step b, by calculating and propagating corresponding check-to-variable messages and updating variable-to-check messages for all variables that received check-to-variable messages in this step; and d. repeating steps a, b and c of calculating values Val i , sorting the calculated set {Val i }, and updating L check node equations until all check nodes are updated by calculating and propagating corresponding check-to-variable messages.
This invention relates to error correction coding, specifically improving the efficiency of soft-input soft-output (SISO) decoding for outer codes in concatenated coding schemes. The problem addressed is the computational complexity and latency in decoding outer codes, particularly when using parity check matrices with a large number of parity check rows (L). The invention optimizes the decoding process by selectively updating check equations in a greedy manner, reducing the number of operations while maintaining decoding accuracy. The method involves selecting a subset of M check nodes (where 1 < M < L) from the parity check matrix for partial updates during each iteration. For the non-updated check nodes (M′ < M), the method calculates a value Val_i for each node, defined as the sum of the two smallest absolute values of variable-to-check messages associated with that node. These values are sorted in decreasing order to prioritize updates for check nodes with the largest Val_i. The method then updates L check node equations according to this sorted order, propagating check-to-variable messages and updating variable-to-check messages for affected variables. This process repeats until all check nodes are updated. The greedy scheduling approach ensures that the most critical check equations are processed first, improving convergence speed and reducing computational overhead.
19. The method of claim 14 , wherein the soft-input soft-output decoding of said outer code operates on log-likelihood ratios of information bits of the convolutional code using message passing, and for at least one variable node, generating check-to-variable messages during an iteration comprises: a. calculating a check-to-variable message Mcv(i,j) from check node i to variable node j; b. identifying two smallest absolute values, Min 1 and Min 2 , in a set of variable-to-check messages Mvc(i, k), where k≠j, excluding the message from variable j to check node i, Mcv(i,j); c. calculating the scaling factor α=1−β ·Min 1 /Min 2 , where β is a non-negative number such that 0<β<1; and d. scaling the check-to-variable message Mcv(i,j) as Mcv(i,j)=α ·Mcv(i,j).
This invention relates to error correction in communication systems, specifically improving soft-input soft-output (SISO) decoding of concatenated codes, such as convolutional codes combined with outer codes like low-density parity-check (LDPC) codes. The problem addressed is enhancing decoding efficiency and accuracy by optimizing message passing in iterative decoding processes. The method involves decoding an outer code (e.g., LDPC) using log-likelihood ratios (LLRs) of information bits from a convolutional code. During each iteration, check-to-variable messages are generated for variable nodes in the decoding graph. For a given check node, the process calculates a check-to-variable message (Mcv(i,j)) from check node i to variable node j. It then identifies the two smallest absolute values (Min1 and Min2) among variable-to-check messages (Mvc(i,k)) received from other variable nodes (k≠j), excluding the message from variable node j to check node i. A scaling factor (α) is computed as α = 1 − β · Min1 / Min2, where β is a non-negative number between 0 and 1. Finally, the check-to-variable message is scaled by this factor (Mcv(i,j) = α · Mcv(i,j)) to improve convergence and reduce decoding errors. This approach refines message passing, leading to more reliable decoding decisions.
20. The method of claim 14 , wherein the soft-input soft-output decoding of said outer code operates on log-likelihood ratios of information bits of the convolutional code using message passing, and for at least one variable node, modifying check-to-variable messages during an iteration comprises: a. calculating variable-to-check messages Mvc(i,j) (n−1) and Mvc(i,j) (n) from variable node j to check node i, respectively, in iterations (n−1) and (n), where n≧2; b. comparing the positive or negative signs of variable-to-check messages Mvc(i,j) (n−1) and Mvc(i,j) (n) calculated in step a; and c. if said signs in step b) are different, generate modified variable-to-check message {tilde over (M)}vc(i,j) (n) according to {tilde over (M)}vc(i,j) (n) =g·Mvc(i,j) (n) +(1−g)·Mvc(i,j) (n−1) , where 0.5<g≦1.
This invention relates to error correction in communication systems, specifically improving soft-input soft-output (SISO) decoding of outer codes in concatenated coding schemes. The problem addressed is enhancing the reliability of decoded information bits in systems using convolutional codes by refining message-passing algorithms during iterative decoding. The method involves decoding an outer code using log-likelihood ratios (LLRs) of information bits from a convolutional code. During iterative decoding, variable-to-check messages are exchanged between variable nodes and check nodes. For at least one variable node, the method modifies check-to-variable messages in each iteration. First, variable-to-check messages from a variable node to a check node are calculated in consecutive iterations. The signs of these messages are then compared. If the signs differ, a modified variable-to-check message is generated as a weighted sum of the current and previous messages, where the weight factor g is between 0.5 and 1. This adjustment improves convergence and accuracy by reducing oscillations in message passing, leading to more reliable decoding. The technique is particularly useful in systems requiring high error correction performance, such as wireless communications or data storage.
21. The method of claim 14 , wherein the outer code is selected from the group consisting of: Reed-Solomon (RS) code and Bose-Chaudhuri-Hocquenghem (BCH) code.
This invention relates to error correction coding in data transmission or storage systems, specifically addressing the need for robust error detection and correction in noisy or unreliable communication channels. The method involves encoding data using a combination of inner and outer error correction codes to enhance reliability. The outer code, which is applied first, is selected from Reed-Solomon (RS) or Bose-Chaudhuri-Hocquenghem (BCH) codes. These codes are well-suited for correcting burst errors and random errors, respectively, providing flexibility in handling different error patterns. The inner code, applied after the outer code, further strengthens error correction by addressing residual errors that the outer code may not fully resolve. The combined approach improves overall system resilience, ensuring data integrity in applications such as wireless communication, optical storage, or deep-space telemetry, where error rates can be high. The selection of the outer code depends on the specific error characteristics of the channel, allowing optimization for performance and efficiency. This method is particularly useful in scenarios where traditional single-layer error correction is insufficient, offering a layered defense against data corruption.
Cooperative Patent Classification codes for this invention. Click any code to explore related patents in that topic.
November 16, 2015
July 4, 2017
Browse 5M+ US patents with plain-English claim translations and AI-generated analysis.